Are Loss Functions All the Same?

DOI10.1162/089976604773135104zbMATH Open1089.68109OpenAlexW2034365297WikidataQ34311744 ScholiaQ34311744MaRDI QIDQ4832479FDOQ4832479

Authors: Lorenzo Rosasco, Michele Piana, Alessandro Verri, Ernesto De Vito, Andrea Caponnetto 

Publication date: 4 January 2005

Published in: Neural Computation (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1162/089976604773135104

Recommendations

• Convexity, Classification, and Risk Bounds
• How to compare different loss functions and their risks
• Statistical behavior and consistency of classification methods based on convex risk minimization.
• Loss functions
• Classification with a reject option using a hinge loss

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• Regularization networks and support vector machines
• Theory of Reproducing Kernels
• On the mathematical foundations of learning
• The covering number in learning theory
• Best choices for regularization parameters in learning theory: on the bias-variance problem.
• Statistical properties and adaptive tuning of support vector machines

Cited In (40)

• Functional linear regression with Huber loss
• Convexity, Classification, and Risk Bounds
• Selection dynamics for deep neural networks
• Analysis of support vector machines regression
• Title not available (Why is that?)
• Analysis of regression algorithms with unbounded sampling
• An investigation for loss functions widely used in machine learning
• Some thoughts about the design of loss functions
• Analysis of loss functions in support vector machines
• Incremental proximal gradient scheme with penalization for constrained composite convex optimization problems
• Levenberg-Marquardt multi-classification using hinge loss function
• A random block-coordinate Douglas-Rachford splitting method with low computational complexity for binary logistic regression
• Bias of homotopic gradient descent for the hinge loss
• Learning rates of kernel-based robust classification
• On the need for structure modelling in sequence prediction
• Genuinely distributed Byzantine machine learning
• Risk-sensitive loss functions for sparse multi-category classification problems
• A statistical learning assessment of Huber regression
• An efficient primal dual prox method for non-smooth optimization
• Fast convergence rates of deep neural networks for classification
• Tensor networks in machine learning
• Loss functions
• A supervised fuzzy measure learning algorithm for combining classifiers
• SVM Soft Margin Classifiers: Linear Programming versus Quadratic Programming
• Good edit similarity learning by loss minimization
• Local Rademacher complexity: sharper risk bounds with and without unlabeled samples
• Classification with a reject option using a hinge loss
• Optimizing predictive precision in imbalanced datasets for actionable revenue change prediction
• Nonasymptotic analysis of robust regression with modified Huber's loss
• Double-well net for image segmentation
• Composite multiclass losses
• A Framework of Learning Through Empirical Gain Maximization
• The C-loss function for pattern classification
• Optimal shrinkage estimation of predictive densities under \(\alpha\)-divergences
• The learning rate of \(l_2\)-coefficient regularized classification with strong loss
• Loss functions for finite sets
• Accelerate stochastic subgradient method by leveraging local growth condition
• How to compare different loss functions and their risks
• Dropout training for SVMs with data augmentation
• A study on L2-loss (squared hinge-loss) multiclass SVM

This page was built for publication: Are Loss Functions All the Same?

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4832479)